DE102008018355A1 - Position measuring device has material measure and sensing head moving against material measure, where material measure has multiple pseudorandom divalent markings with constant measuring partition - Google Patents

Abstract

The position measuring device (10) has a material measure (11) and a sensing head (13) moving against the material measure. The material measure has multiple pseudorandom divalent markings (12) with a constant measuring partition (t). Multiple sensors are provided in the sensing head, which are arranged so that they sense multiple divalent markings in a position of the position measuring device. A sensing signal is guided indirectly to both the absolute detector and the interpolator.

Description

The The invention relates to a position measuring device according to the Preamble of claim 1.

From the DE 102 34 744 A1 a position measuring device is known. According to the local 4 is a material measure 10 provided, which is scanned by a scanning head. The measuring standard comprises a multiplicity of markings arranged in a row in the form of magnetic north and south poles, so that each mark can assume two possible values. The north and south poles are distributed pseudorandomly with a constant measurement graduation, ie, the bit code of a predetermined number of juxtaposed markers is assigned a unique absolute position of the scanning head relative to the material measure. Provided in the scanning head are a plurality of sensors SC0 to SC9 and SF0 to SF6, which are arranged so that they can simultaneously scan a plurality of bivalent marks in one position of the position measuring devices. Said sensors are designed as Hall sensors, accordingly their scanning signals change continuously when the position measuring device changes its position. The scanning signals of the sensors SC0 to SC9 become an absolute decoder 14 ; 16 ; 20 supplied to determine an absolute position with respect to the measurement graduation. Next is an interpolator 24 ; 26 ; 28 provided, which determines from the scanning signals of the sensors SF0 to SF5 an intermediate position with respect to the measuring graduation. The total position value composed of the absolute position and the intermediate position is therefore more accurate than the measurement graduation of the material measure.

Of the Disadvantage of this position measuring device is that very Many sensors are required for position determination are. The object of the invention is accordingly, the number to reduce the sensors.

To solve this problem, it is proposed that at least one scanning signal can be fed at least indirectly both to the absolute decoder and to the interpolator. Accordingly, for the absolute decoder and the interpolator not separate sensors as in the DE 102 34 744 A1 but uses common sensors. As a result, fewer sensors are required for position determination.

A particularly small number of sensors results if all the scanning signals can be fed at least indirectly both to the absolute decoder and to the interpolator. In an arrangement of the sensors as in the DE 102 34 744 A1 2, where the sensor pitch is equal to the measurement pitch, the required number of sensors is the code width of the pseudorandom code multiplied by two.

The distance s of the sensors may deviate from the measuring graduation t, satisfying the condition n · s ≥ m · t, where n is the number of sensors and m is the code width of the pseudo-random code and where n> m. By this configuration, the number of required sensors can be further reduced. This embodiment is based on the same functional principle, which is also used in the vernier scale of a mechanical calliper. The smallest possible number of sensors results if the condition n = m + 1 with n · s = m · t is satisfied. The deviation of the distance of the sensors from the measuring graduation ensures that the various sensors each have a different relative position relative to the individual opposing markings, so that a mathematically meaningful evaluation of the scanning signals is possible. In the known embodiment according to the DE 102 34 744 A1 This problem was taken into account by two separate sensor combs which are offset by half the measurement graduation.

It At least one analog-to-digital converter may be provided to to convert all scanning signals into a digital measuring vector, wherein at least one digital multiplier unit is provided which is set up so that it sets the measurement vector with several Can multiply matrices to obtain multiple code vectors. In this way, the position measuring device can be particularly simple are executed, since only addition and multiplication operations required are. The mentioned mathematical operations can be largely be executed in parallel, whereby the output of the position measurement value can be done with a very low time delay.

Of the Interpolator can be set up to be codevectors choose the one who has the slightest deviation to a two-valued code to the relative intermediate position to determine the measurement graduation. By this configuration the interpolation can be realized in a particularly fast manner, since the deviations of the different code vectors to a bivalent one Code can be easily executed in parallel time.

The interpolator may be arranged to determine the deviation of a code vector to a bivalent code by calculating the magnitude difference for each code vector element to be 0 and 1, calculating a predetermined power of the minimum of the two magnitude differences, and wherein summed up. Thus, the deviation of a code vector from a two-valued code can particularly simple and therefore inexpensive to be implemented arithmetic operations are performed. The given power of the minimum of the two magnitude differences is preferably the second power, since this is particularly easy to calculate.

Of the The codevector selected by the interpolator may be the absolute decoder be fed. The absolute decoder can work in a familiar way work with a table containing every possible bitcode assigns the corresponding absolute position. The selected Codevector that comes closest to a two-valued code is for access to said table on the said two-valued Code rounded. The described absolute decoder is very simple built and therefore inexpensive to manufacture. The mentioned processing steps can be carried out very quickly, whereby the output of the position measurement value with a particularly low Time delay can be done.

The Markings of the material measure can of high and low magnetic permeability sections or sections of high and low electrical conductivity are formed, wherein the sensors electrical coils, preferably Planar coils, are. Such dimensional standards can be made particularly inexpensive. The measuring standard may be, for example, a thin sheet metal strip, into which a Variety of similar, rectangular openings etched has been. The two possible values of the mark result through the presence or absence of a breakthrough. But it is also possible, the guide rail of a linear roller bearing to use as a material measure, with the markings be formed by recesses in the guide rail. The recesses can, for example by means of spark erosion incorporated into the hardened steel guide rail become. Due to the one-piece design of the material measure a particularly high measuring accuracy can be achieved.

The proposed inductive scanning has the advantage of a particular high resistance to environmental influences, such as contamination of the material measure. The hereby used coils, in particular the cost planar coils are to carry out with a particularly high number of turns so that a particularly strong sensor signal is generated. In this context It should be noted that the distance between the material measure and the coils usually of the same order of magnitude how the measuring graduation of the material measure lies. This circumstance leads to a relatively weak sensor signal, which only a little above the unavoidable noise voltage the coils used is. Accordingly, the maximization the number of turns of the coils particularly important, since this is the Signal strength increases much more than that Noise voltage of the sensor coils.

The Invention will be described below with reference to the accompanying drawings explained in more detail. It shows:

1 a rough schematic representation of a position measuring device according to the invention and

2 a block diagram of the position measuring device according to 1 ,

In 1 is a position measuring device according to the invention in general with 10 designated. The position measuring device 10 consists of a material measure 11 attached to the guide rail 14 a linear roller bearing is provided. At the guide rail 14 is a carriage 15 over several rows of (not shown) rolling elements in the longitudinal direction 16 movably guided. At one end of the carriage 15 is a scanning head 13 provided in which the sensors for scanning the material measure 11 are arranged. By evaluating the scanning signals of the sensors, the absolute position of the guide carriage 15 opposite the guide rail 14 certainly.

The measuring standard 11 is characterized by a variety of divalent markings 12 formed on the guide rail made of hardened bearing steel. The marks 12 are pseudo-random with a constant measurement pitch t of 1.0 mm along the longitudinal direction 16 the guide rail 14 arranged distributed. The two possible values of a marker 12 are represented by the absence or presence of a recess in the guide rail, wherein the recesses by means of spark erosion with high precision in the guide rail 14 are incorporated. The entire material measure is tightly covered with a (not shown) thin strip of sheet steel, so that the recesses can not be contaminated by foreign substances, which could change their electromagnetic properties.

The pseudorandomly arranged bivalent markers 12 have the property that a predetermined number (code width) of adjacent bivalent marks form a bit code uniquely identifying the absolute position of this bit code on the scale.

2 shows a block diagram of the evaluation circuit in the scanning 13 is arranged. The evaluation circuit comprises several sensors 30 at a predetermined constant distance s along the longitudinal direction 16 arranged distributed are, wherein the distance s deviates from the measuring graduation t. In the present block diagram, the code width of the material measure is four, wherein the sensor distance s is chosen so that over a distance of four measuring graduations t five sensors are arranged uniformly distributed. It goes without saying that a substantially larger code width is required in a practically usable position measuring device, so that a sufficiently large number of absolute positions can be distinguished. The number of absolute positions to be distinguished results from the measuring graduation t and the length of the material measure 11 or the guide rail.

The sensors 30 are designed as planar coils, which are applied by means of a photochemical etching process on a substantially rigid planar substrate made of silicon, metal or plastic. A sensor may optionally be a single coil with a single winding direction or a series-connected coil pair with the opposite direction of winding. The second so-called differential coil arrangement is chosen if a high insensitivity of the position measuring system with respect to external disturbances is desired. The first coil arrangement is preferred if a very small measuring graduation t is aimed for in order to increase the measuring accuracy. The planar coils 30 are aligned parallel to the surface of the measuring scale, wherein the distance to this surface corresponds approximately to the measuring graduation t.

Further, in the scanning head at least one (not shown) field generating coil is provided, which generates an alternating electromagnetic field, all the sensors 30 interspersed. This electromagnetic alternating field is modulated by the recesses in the guide rail, whereby, depending on the presence or absence of a recess different AC voltages in the sensor coils 30 be induced. The scanning signals 52 in the form of said alternating voltages change continuously with the position of the guide carriage relative to the guide rail.

The scanning signals 52 all sensors 30 are each by an analog-to-digital converter 33 digitized. The use of several analog-to-digital converters operating in parallel 33 is preferred over a single multiplexed analog-to-digital converter because of greater travel speeds of the position measuring device 10 possible are. In addition, the exact simultaneous digitization of all scanning signals increases 52 the measuring accuracy when the position measuring device 10 is moved quickly. The analog-to-digital conversion becomes synchronous to the electromagnetic alternating field, which is the scanning signals 52 generated, performed. Consequently, the obtained digital values represent the oscillation amplitude of the AC voltages induced in the sensor coils. One speaks in this context of a synchronous demodulation.

All digital values used when digitizing the scanning signals 52 be obtained at a given time, become a digital measurement vector 53 summarized. The digital measurement vectors 53 become multiple parallel multiplier units 34 fed to the measuring vector 53 in each case multiply by a predefinable matrix of digital values to form a code vector 54 to investigate. Each of these matrices is assigned to one of the intermediate position to be determined of the position measuring device with respect to the measuring graduation. The codevector 54 corresponds to that under the sensors 30 existing bivalent bit code of the material measure 11 , provided that the position measuring device is located exactly in the intermediate position, which is assigned to the relevant matrix.

The Calculation of the code vector or bit code of the material measure from the measurement vector can be determined with the help of a matrix multiplication perform because the proposed position measuring device has a linear system behavior, i. H. the measurement vector, the For example, results in a codevector of (1, 1, 0, 1) leaves to calculate from the sum of the measuring vectors, which result from the Code vectors (1, 0, 0, 0), (0, 1, 0, 0) and (0, 0, 0, 1) result. The latter vectors are called unit vectors, since they only contain a single 1-bit. The matrices mentioned can be Consequently, determine by the measurement vectors to the various Unit vectors are determined experimentally. This will be a Measuring standard used only a single one 1-mark, so only a single recess, wherein the Scanning head in different positions with respect to the 1-mark the different unit vectors and the different ones correspond to intermediate positions to be determined. In the mentioned positions the respective measuring vectors are determined. From these measurements in conjunction with the associated unit vectors results for each sought matrix a linear system of equations, the can be solved easily. It should be noted that this equation system has more equations than unknown, since according to the invention, the number of sensors larger as the code width is. In the presence of an exactly linear system behavior the extra equations are linearly dependent to the other equations. In the case of one easily nonlinear system behavior becomes the solution of the System of equations replaced by an optimization task in which the matrix is sought, which the said equation system with solves the slightest mistake.

The codevectors 54 are each a deviation determination unit 35 fed the distance 55 of the code vector 54 determined by a two-valued bit code. This calculation step is based on the consideration that the above matrix calculation only supplies a divalent codevector, that is to say a codevector which has only the values 0 and 1, when the position-measuring device is in the intermediate position assigned to the matrix. In all other cases, the code vector of 0 or 1 will have different values, the deviation being greater the farther the position measuring device is away from the intermediate position under consideration. The mentioned deviation 55 of a codevector to a bivalent code is determined by calculating for each codevector element the magnitude difference to 0 and to 1, calculating the square of the minimum of the two magnitude differences and summing up said squares.

The deviations 55 become a selection unit 36 fed, which is the smallest deviation 55 selects and from this the intermediate position 55 determined. This intermediate position 55 is on the one hand as part of the total position value 56 outputted and on the other hand fed to the absolute decoder, so this on the basis of the intermediate position 55 the most suitable code vector for further processing 54 can choose. The position measuring device preferably operates in the dual number system, for which reason the number of intermediate positions is preferably a power of two. The intermediate position 55 in this case corresponds to the least significant bits of the total position value. The interpolator 32 in the sense of the independent claim is determined by the multiplier units 34 , the variance calculation units 35 and the selection unit 36 educated.

The most significant bits of the total position value 56 or the absolute position 51 is from the absolute decoder 31 determined by the basis of the intermediate position 50 selected codevector 54 is rounded to a divalent codevector, ie all code elements of the selected codevector 54 are rounded to either 0 or 1. As a result, a finite number of possible bit codes is contained, to which the absolute position sought is displayed via a finite table 51 is assigned. It should be noted that the absolute decoder 31 the scanning signals 52 in the sense of the independent claim indirectly via the code vector 54 be supplied.

t

measuring graduation

s

distance the sensors

10

Position measuring device

11

Measuring standard

12

mark

13

scan head

14

guide rail

15

carriages

16

longitudinal direction

30

sensor

31

absolutely decoder

32

interpolator

33

Analog-to-digital converter

34

multiplier

35

Variance calculation unit

36

selector

50

intermediate position

51

absolute position

52

sampling

53

measurement vector

54

code vector

55

deviation

56

Total position value

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited patent literature

• DE 10234744 A1 [0002, 0004, 0005, 0006]

Claims (9)

Position measuring device ( 10 ) with a material measure ( 11 ) and one opposite the material measure ( 11 ) movable scanning head ( 13 ) for scanning the material measure ( 11 ), wherein the material measure ( 11 ) a plurality of pseudorandom divalent labels arranged in a row ( 12 ) with a constant measuring graduation (t), wherein in the scanning head ( 13 ) several sensors ( 30 ) are arranged, which are arranged so that they in a position of the position measuring device ( 10 ) several bivalent markers ( 12 ), whereby the sensors ( 30 ) Scanning signals ( 52 ) can be output, which in a position change of the position measuring device ( 10 ), whereby the scanning signals are sent to an absolute decoder ( 31 ) for determining an absolute position ( 51 ) with respect to the measurement graduation (t) and an interpolator ( 52 ) for determining a relative intermediate position ( 50 ) with respect to the measuring graduation (t), characterized in that at least one scanning signal ( 52 ) at least indirectly both the absolute decoder ( 51 ) as well as the interpolator ( 32 ) can be fed. Position measuring device according to one of the preceding claims, characterized in that all scanning signals ( 52 ) at least indirectly both the absolute decoder ( 31 ) as well as the interpolator ( 32 ) can be supplied. Position measuring device according to claim 1, characterized in that the distance s of the sensors ( 30 ) differs from the measuring graduation t, satisfying the condition n · s ≥ m · t, where n is the number of sensors and m is the code width of the pseudo-random code and where n> m. Position measuring device according to claim 3, characterized characterized in that the condition n = m + 1 applies. Position measuring device according to one of the preceding claims, characterized in that at least one analog-to-digital converter ( 33 ) is provided to all the scanning signals ( 52 ) into a digital measurement vector ( 53 ), wherein at least one digital multiplication unit ( 34 ) arranged to receive the measurement vector ( 53 ) can multiply with several predetermined matrices to form multiple code vectors ( 54 ) to obtain. Position measuring device according to claim 5, characterized in that the interpolator ( 32 ) is arranged to be selected from the codevectors ( 54 ) can select the one with the least deviation ( 55 ) to a bivalent code to determine the relative intermediate position with respect to the measurement pitch (t). Position measuring device according to claim 6, characterized in that the interpolator ( 32 ) is arranged to detect the deviation of a code vector ( 54 ) to a bivalent code by calculating, for each code vector element, the magnitude difference to 0 and 1, calculating a predetermined power of the minimum of the two magnitude differences and summing up said powers. Position measuring device according to claim 6 or 7, characterized in that that of the interpolator ( 32 ) selected codevector ( 54 ) the absolute decoder ( 31 ) can be fed. Position measuring device according to one of the preceding claims, characterized in that the markings ( 12 ) of the material measure ( 11 ) are formed of sections of high and low magnetic permeability or of sections of high and low electrical conductivity, wherein the sensors ( 30 ) Electric coils, preferably planar coils, are.